Downloaded from http://jramc.bmj.com/ on September 2, 2015 - Published by group.bmj.com
Original article
Screening for latent tuberculosis and gastrointestinal parasite infections in Gurkha recruits: research driving policy change Matthew K O’Shea,1,5 T E Fletcher,2,5 D Tupper,3 D Ross,4 D Wilson5 1
Nuffield Department of Medicine, The Jenner Institute, University of Oxford, Oxford, UK 2 Liverpool School of Tropical Medicine, Liverpool, UK 3 Medical Centre, Vimy Barracks, Catterick, UK 4 Army Health Unit, Former Army Staff College, Camberley, UK 5 Royal Centre for Defence Medicine (Academia and Research), Birmingham, UK Correspondence to Col Duncan Wilson, Royal Centre for Defence Medicine (Academia and Research), Medical Directorate, Joint Medical Command, ICT Building, Birmingham Research Park, Birmingham B15 2SQ, UK; [email protected] Received 31 January 2014 Accepted 5 February 2014 Published Online First 7 March 2014
To cite: O’Shea MK, Fletcher TE, Tupper D, et al. J R Army Med Corps 2014;160:180–182. 180
ABSTRACT Nepalese Gurkha soldiers are recruited from a country endemic for a number of infectious diseases, including tuberculosis and gastrointestinal parasites. This article describes a prospective cohort study which investigated screening strategies for these infections among Gurkha recruits arriving in the UK to begin basic training. Several recommendations were made as a result of the study which were supported for early implementation and subsequently fully adopted. Military screening and treatment policies have been directly influenced by this research which also has translational application to similar migrant civilian populations.
BACKGROUND Nepalese Gurkha soldiers have served with distinction in the British army for almost 200 years, but as early as 1908 their propensity to develop tuberculosis (TB) infection was recognised, particularly during their first year of service.1 In the last significant review of pulmonary TB among serving Gurkhas, published in 1964, the annual incidence of new cases was estimated at 3 per 1000.1 The study concluded that ‘as long as Gurkhas continue to serve with the British Army their particular proneness to develop tuberculosis must be recognised and taken account of ’. Today, while the number of Gurkha soldiers has reduced, the problems associated with recruiting individuals from a region endemic for TB persist—the estimated incidence in Nepal is 163 cases per 100 000 population.2 The recognition of TB infection in such a migrant population has important considerations for the wider civilian community. In the UK, TB notifications have increased continuously over the past 30 years and of the 9000 new cases notified in 2011, 74% occurred among foreign-born individuals.3 Strategies for the effective control and elimination of TB in low-incidence countries, such as the UK, depend on the prompt identification and management of active disease, and also those individuals with latent TB infection (LTBI) who are at risk of reactivation, progression to disease and infectiousness.4 5 The diagnosis of LTBI remains challenging and, until recently, the century-old tuberculin skin test (TST) had been the diagnostic method of choice. While cheap and relatively easy to perform, the limitations of TST are well documented. These include placement and inter-reader variability, inaccurate interpretation of results over variable timescales, false positive results due to immune sensitisation,
Key messages ▸ Nepalese Gurkha soldiers serving in the British Army are recruited from a region with a high incidence of tuberculosis (TB) and soiltransmitted helminthiasis. ▸ We assessed screening strategies for latent TB infection in a group of Gurkha recruits in the UK and the impact of gastrointestinal parasite infection on test results. ▸ A single-step T-SPOT.TB IGRA and detailed stool analysis, followed by parasite eradication and IGRA re-testing was the most sensitive, specific and efficient strategy in this group. ▸ Study findings have directly influenced DMS policy and have implications for screening strategies in similar civilian migrant populations.
patient inconvenience (having to return in 48–72 h for assessment), and low sensitivity and specificity.6 Early secretory antigenic target-6 (ESAT-6), culture filtrate protein-10 (CFP-10) and TB7.7 are antigenic gene products that elicit pronounced Th1 cellular immune responses in Mycobacterium tuberculosis infection, and have subsequently been used in novel T cell-based interferon-γ (IFN-γ) release assays (IGRAs).7 The suggested advantages of IGRAs over the TST include a single patient visit for blood sampling, availability of results within 24 h, no confounding effect of prior vaccination with BCG or previous exposure to non-TB mycobacteria, and no boosting or sensitisation effect which preserves the option to repeat the test in the same patient on several occasions.8 However, IGRAs are not a diagnostic panacea for LTBI, and problems with subject variability, reproducibility of results, performance in serial testing and discordance between assays have been reported.9 10 For the past decade two commercial IGRAs have been available. The QuantiFERON-TB Gold In-Tube (QFT-GIT; Qiagen, Venlo, The Netherlands) is a whole blood ELISA, whereas the T-SPOT.TB assay (Oxford Immunotec, Abingdon, UK) is an enzymelinked immunospot (ELISPOT) technique using peripheral blood mononuclear cells. The former assay measures IFN-γ concentrations, while the latter enumerates IFN-γ producing T lymphocytes, both in response to stimulation with M tuberculosis-specific antigens. There are few data comparing the TSTwith the commercially available IGRAs in migrants.
O’Shea MK, et al. J R Army Med Corps 2014;160:180–182. doi:10.1136/jramc-2014-000259
Downloaded from http://jramc.bmj.com/ on September 2, 2015 - Published by group.bmj.com
Original article Screening Gurkha recruits arriving in the UK for TB presents the challenges outlined above and an opportunity to study different strategies in a controlled and easily followed population. Each year there are 15 000–20 000 applicants of whom approximately 1% are recruited and arrive every January at the Infantry Training Centre (ITC), Catterick Garrison, to commence a 10-month programme of intensive phase-1 training. Part of the selection process includes basic medical screening in Nepal. On arrival at ITC, a further medical review is conducted and formal screening for LTBI is undertaken. Prior to 2012, this employed a two-step strategy of a TST followed by an IGRA (QFT-GIT) in those with a positive TST result (induration of ≥6 mm in BCG-naive, or ≥15 mm in BCG-vaccinated individuals) in accordance with national guidelines.11 After secondary care review to exclude active disease and contraindications, recruits with a positive IGRA and normal chest radiography were offered LTBI treatment consisting of a 3-month regimen of daily fixed-dose combination rifampicin and isoniazid. A one-step strategy could result in considerable benefit and was felt worthy of assessment. Together with a high TB incidence, soil-transmitted helminthiasis is responsible for a significant burden of infection in Nepal. The last published epidemiological study of helminth infection in British Gurkha recruits from 1978, reported hookworm, whipworm and roundworm infestations in 100%, 60% and 60% of recruits, respectively.12 The policy employed at ITC was to empirically treat Gurkha recruits with a single dose of albendazole upon arrival in the UK. No such strategy currently exists for equivalent civilian immigrants arriving in the UK from high disease-burden settings, by contrast with other national policies, such as in the USA and Australia. A new stool-based helminth prevalence study, together with serology for Strongyloides stercoralis infection, would be valuable in assessing the relevance of such a policy, while also presenting the opportunity to further assess the likely impact of helminth infection on the host immune response to TB, and therefore the TST and IGRA tests. This has been a focus of increasing research interest recently.13 There have been conflicting reports concerning the impact of active helminth infection on TST responses. Among asymptomatic children with a history of exposure to M tuberculosis and concomitant parasitic infections, some investigators have reported a negative association between TST positivity and coinfection.13 However, of the IGRAs, only QFT-based assays have been assessed, and an association between helminth infections and indeterminate QFT results has been demonstrated. Previous
studies have been limited to the assessment of paediatric populations, diagnosing helminth infection on the basis of positive serology and/or peripheral eosinophilia, and little assessment of the possible association between helminth or protozoal infections with negative IGRA responses.
THE STUDY A research protocol was developed with the full support of Surgeon General, Joint Medical Command, Brigade of Gurkhas Headquarters, the Gurkha Welfare Trust and the Gurkha Company, medical centre and Headquarters ITC Catterick. Ethical approval was granted by the Ministry of Defence Research ethics committee. The prospective cohort study was undertaken from February to September 2012 with the aims: ▸ To directly compare the TST with both commercially available IGRAs for the diagnosis of LTBI in Gurkha recruits. ▸ To assess the prevalence of gastrointestinal parasite infections in the recruits. ▸ To investigate the effect of gastrointestinal parasite infections on IGRA results. Detailed methodology of the study and full results will be published elsewhere. Briefly, TST, IGRAs and assessment of stool samples for parasites were performed at the commencement of the study, with IGRA repeated at 7 and 200 days. One-hundred and sixty-six Gurkha recruits participated in the study and a complete dataset was collected at all time points. Of these recruits, 17.5% had evidence of LTBI. Analysis of the TST and IGRA results concluded that a single-step IGRA, and specifically a T-SPOT.TB test, was the most sensitive, specific and efficient strategy for diagnosing LTBI in this population. Gastrointestinal parasites were identified among 27% of the recruits, including a number with multiple pathogens, indicating that the existing policy of empirical albendazole treatment was inadequate for complete gastrointestinal parasite eradication. It was also evident that gastrointestinal parasite infection is likely to attenuate responses to LTBI testing leading to the possibility of false negative results.
IMPACT ON MILITARY POLICY The results of this study, with a 100% cohort follow-up, were unequivocal. T-SPOT.TB provided more consistent results throughout the study and was affected less by TST and LTBI
Figure 1 Latent tuberculosis infection (LTBI) screening strategies. 1 SS=Strongyloides serology. 2 GI=gastrointestinal. 3 Chemoprophylaxis if
Original article
Screening for latent tuberculosis and gastrointestinal parasite infections in Gurkha recruits: research driving policy change Matthew K O’Shea,1,5 T E Fletcher,2,5 D Tupper,3 D Ross,4 D Wilson5 1
Nuffield Department of Medicine, The Jenner Institute, University of Oxford, Oxford, UK 2 Liverpool School of Tropical Medicine, Liverpool, UK 3 Medical Centre, Vimy Barracks, Catterick, UK 4 Army Health Unit, Former Army Staff College, Camberley, UK 5 Royal Centre for Defence Medicine (Academia and Research), Birmingham, UK Correspondence to Col Duncan Wilson, Royal Centre for Defence Medicine (Academia and Research), Medical Directorate, Joint Medical Command, ICT Building, Birmingham Research Park, Birmingham B15 2SQ, UK; [email protected] Received 31 January 2014 Accepted 5 February 2014 Published Online First 7 March 2014
To cite: O’Shea MK, Fletcher TE, Tupper D, et al. J R Army Med Corps 2014;160:180–182. 180
ABSTRACT Nepalese Gurkha soldiers are recruited from a country endemic for a number of infectious diseases, including tuberculosis and gastrointestinal parasites. This article describes a prospective cohort study which investigated screening strategies for these infections among Gurkha recruits arriving in the UK to begin basic training. Several recommendations were made as a result of the study which were supported for early implementation and subsequently fully adopted. Military screening and treatment policies have been directly influenced by this research which also has translational application to similar migrant civilian populations.
BACKGROUND Nepalese Gurkha soldiers have served with distinction in the British army for almost 200 years, but as early as 1908 their propensity to develop tuberculosis (TB) infection was recognised, particularly during their first year of service.1 In the last significant review of pulmonary TB among serving Gurkhas, published in 1964, the annual incidence of new cases was estimated at 3 per 1000.1 The study concluded that ‘as long as Gurkhas continue to serve with the British Army their particular proneness to develop tuberculosis must be recognised and taken account of ’. Today, while the number of Gurkha soldiers has reduced, the problems associated with recruiting individuals from a region endemic for TB persist—the estimated incidence in Nepal is 163 cases per 100 000 population.2 The recognition of TB infection in such a migrant population has important considerations for the wider civilian community. In the UK, TB notifications have increased continuously over the past 30 years and of the 9000 new cases notified in 2011, 74% occurred among foreign-born individuals.3 Strategies for the effective control and elimination of TB in low-incidence countries, such as the UK, depend on the prompt identification and management of active disease, and also those individuals with latent TB infection (LTBI) who are at risk of reactivation, progression to disease and infectiousness.4 5 The diagnosis of LTBI remains challenging and, until recently, the century-old tuberculin skin test (TST) had been the diagnostic method of choice. While cheap and relatively easy to perform, the limitations of TST are well documented. These include placement and inter-reader variability, inaccurate interpretation of results over variable timescales, false positive results due to immune sensitisation,
Key messages ▸ Nepalese Gurkha soldiers serving in the British Army are recruited from a region with a high incidence of tuberculosis (TB) and soiltransmitted helminthiasis. ▸ We assessed screening strategies for latent TB infection in a group of Gurkha recruits in the UK and the impact of gastrointestinal parasite infection on test results. ▸ A single-step T-SPOT.TB IGRA and detailed stool analysis, followed by parasite eradication and IGRA re-testing was the most sensitive, specific and efficient strategy in this group. ▸ Study findings have directly influenced DMS policy and have implications for screening strategies in similar civilian migrant populations.
patient inconvenience (having to return in 48–72 h for assessment), and low sensitivity and specificity.6 Early secretory antigenic target-6 (ESAT-6), culture filtrate protein-10 (CFP-10) and TB7.7 are antigenic gene products that elicit pronounced Th1 cellular immune responses in Mycobacterium tuberculosis infection, and have subsequently been used in novel T cell-based interferon-γ (IFN-γ) release assays (IGRAs).7 The suggested advantages of IGRAs over the TST include a single patient visit for blood sampling, availability of results within 24 h, no confounding effect of prior vaccination with BCG or previous exposure to non-TB mycobacteria, and no boosting or sensitisation effect which preserves the option to repeat the test in the same patient on several occasions.8 However, IGRAs are not a diagnostic panacea for LTBI, and problems with subject variability, reproducibility of results, performance in serial testing and discordance between assays have been reported.9 10 For the past decade two commercial IGRAs have been available. The QuantiFERON-TB Gold In-Tube (QFT-GIT; Qiagen, Venlo, The Netherlands) is a whole blood ELISA, whereas the T-SPOT.TB assay (Oxford Immunotec, Abingdon, UK) is an enzymelinked immunospot (ELISPOT) technique using peripheral blood mononuclear cells. The former assay measures IFN-γ concentrations, while the latter enumerates IFN-γ producing T lymphocytes, both in response to stimulation with M tuberculosis-specific antigens. There are few data comparing the TSTwith the commercially available IGRAs in migrants.
O’Shea MK, et al. J R Army Med Corps 2014;160:180–182. doi:10.1136/jramc-2014-000259
Downloaded from http://jramc.bmj.com/ on September 2, 2015 - Published by group.bmj.com
Original article Screening Gurkha recruits arriving in the UK for TB presents the challenges outlined above and an opportunity to study different strategies in a controlled and easily followed population. Each year there are 15 000–20 000 applicants of whom approximately 1% are recruited and arrive every January at the Infantry Training Centre (ITC), Catterick Garrison, to commence a 10-month programme of intensive phase-1 training. Part of the selection process includes basic medical screening in Nepal. On arrival at ITC, a further medical review is conducted and formal screening for LTBI is undertaken. Prior to 2012, this employed a two-step strategy of a TST followed by an IGRA (QFT-GIT) in those with a positive TST result (induration of ≥6 mm in BCG-naive, or ≥15 mm in BCG-vaccinated individuals) in accordance with national guidelines.11 After secondary care review to exclude active disease and contraindications, recruits with a positive IGRA and normal chest radiography were offered LTBI treatment consisting of a 3-month regimen of daily fixed-dose combination rifampicin and isoniazid. A one-step strategy could result in considerable benefit and was felt worthy of assessment. Together with a high TB incidence, soil-transmitted helminthiasis is responsible for a significant burden of infection in Nepal. The last published epidemiological study of helminth infection in British Gurkha recruits from 1978, reported hookworm, whipworm and roundworm infestations in 100%, 60% and 60% of recruits, respectively.12 The policy employed at ITC was to empirically treat Gurkha recruits with a single dose of albendazole upon arrival in the UK. No such strategy currently exists for equivalent civilian immigrants arriving in the UK from high disease-burden settings, by contrast with other national policies, such as in the USA and Australia. A new stool-based helminth prevalence study, together with serology for Strongyloides stercoralis infection, would be valuable in assessing the relevance of such a policy, while also presenting the opportunity to further assess the likely impact of helminth infection on the host immune response to TB, and therefore the TST and IGRA tests. This has been a focus of increasing research interest recently.13 There have been conflicting reports concerning the impact of active helminth infection on TST responses. Among asymptomatic children with a history of exposure to M tuberculosis and concomitant parasitic infections, some investigators have reported a negative association between TST positivity and coinfection.13 However, of the IGRAs, only QFT-based assays have been assessed, and an association between helminth infections and indeterminate QFT results has been demonstrated. Previous
studies have been limited to the assessment of paediatric populations, diagnosing helminth infection on the basis of positive serology and/or peripheral eosinophilia, and little assessment of the possible association between helminth or protozoal infections with negative IGRA responses.
THE STUDY A research protocol was developed with the full support of Surgeon General, Joint Medical Command, Brigade of Gurkhas Headquarters, the Gurkha Welfare Trust and the Gurkha Company, medical centre and Headquarters ITC Catterick. Ethical approval was granted by the Ministry of Defence Research ethics committee. The prospective cohort study was undertaken from February to September 2012 with the aims: ▸ To directly compare the TST with both commercially available IGRAs for the diagnosis of LTBI in Gurkha recruits. ▸ To assess the prevalence of gastrointestinal parasite infections in the recruits. ▸ To investigate the effect of gastrointestinal parasite infections on IGRA results. Detailed methodology of the study and full results will be published elsewhere. Briefly, TST, IGRAs and assessment of stool samples for parasites were performed at the commencement of the study, with IGRA repeated at 7 and 200 days. One-hundred and sixty-six Gurkha recruits participated in the study and a complete dataset was collected at all time points. Of these recruits, 17.5% had evidence of LTBI. Analysis of the TST and IGRA results concluded that a single-step IGRA, and specifically a T-SPOT.TB test, was the most sensitive, specific and efficient strategy for diagnosing LTBI in this population. Gastrointestinal parasites were identified among 27% of the recruits, including a number with multiple pathogens, indicating that the existing policy of empirical albendazole treatment was inadequate for complete gastrointestinal parasite eradication. It was also evident that gastrointestinal parasite infection is likely to attenuate responses to LTBI testing leading to the possibility of false negative results.
IMPACT ON MILITARY POLICY The results of this study, with a 100% cohort follow-up, were unequivocal. T-SPOT.TB provided more consistent results throughout the study and was affected less by TST and LTBI
Figure 1 Latent tuberculosis infection (LTBI) screening strategies. 1 SS=Strongyloides serology. 2 GI=gastrointestinal. 3 Chemoprophylaxis if
